pi-condense 2.4.0 → 2.4.1

package/CHANGELOG.md

@@ -7,7 +7,9 @@ Published to npm as [`pi-condense`](https://www.npmjs.com/package/pi-condense) (
 Pushing a `vX.Y.Z` tag triggers `.github/workflows/release.yml`, which runs the tests and
 publishes via OIDC trusted publishing. See `.agents/skills/release/SKILL.md`.
 
-## [Unreleased]
+## [2.4.1] - 2026-07-30
+
+- **Fix 421 Misdirected Request for GitHub Copilot business/enterprise seats.** The summarizer called pi-ai's `stream()` with the static model definition, whose shipped `baseUrl` pins the individual Copilot host (`api.individual.githubcopilot.com`); business/enterprise tokens are rejected there with `421 Misdirected Request`, so any `github-copilot/*` `summarizerModel` failed on those seats. `runOnce` now mirrors the main agent loop (pi's `model-runtime`): it resolves provider auth via `ctx.modelRegistry.getProviderAuth(model.provider)` and, when the resolved auth carries a seat-specific `baseUrl`, rebases the model onto it before streaming. Also bumps `@earendil-works/*` devDependencies to `^0.83.0` (needed for `getProviderAuth` on the extension-facing `ModelRegistry`) and imports `stream` from `@earendil-works/pi-ai/compat` (its export moved off the root in current pi-ai).
 
 ## [2.4.0] - 2026-07-09
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-condense",
-  "version": "2.4.0",
+  "version": "2.4.1",
   "description": "Pi coding-agent extension that summarizes completed tool-call batches, replaces raw outputs with short stubs, compresses closed tool-call chains, and recovers any original on demand via context_tree_query.",
   "author": "Jacek Juraszek",
   "license": "MIT",
@@ -57,9 +57,9 @@
     "@sinclair/typebox": "*"
   },
   "devDependencies": {
-    "@earendil-works/pi-ai": "^0.78.0",
-    "@earendil-works/pi-coding-agent": "^0.78.0",
-    "@earendil-works/pi-tui": "^0.78.0",
+    "@earendil-works/pi-ai": "^0.83.0",
+    "@earendil-works/pi-coding-agent": "^0.83.0",
+    "@earendil-works/pi-tui": "^0.83.0",
     "@sinclair/typebox": "^0.34.49"
   }
 }

package/src/summarizer-wiring.test.ts

@@ -5,7 +5,7 @@ import { describe, it, expect, mock } from "bun:test";
 let streamImpl: (model: any, input?: any, opts?: any) => any = () => {
   throw new Error("streamImpl not set");
 };
-mock.module("@earendil-works/pi-ai", () => ({
+mock.module("@earendil-works/pi-ai/compat", () => ({
   stream: (...args: any[]) => streamImpl(...args),
 }));
 
@@ -111,6 +111,7 @@ function makeCtx(notes: Note[], sessionModel: any = SESSION, primaryModel: any =
     modelRegistry: {
       find: () => primaryModel,
       getApiKeyAndHeaders: async () => ({ ok: true, apiKey: "k", headers: {} }),
+      getProviderAuth: async () => undefined,
     },
     ui: { notify: (msg: string, level: string) => notes.push({ msg, level }) },
   } as any;

package/src/summarizer.ts

@@ -1,4 +1,4 @@
-import { stream } from "@earendil-works/pi-ai";
+import { stream } from "@earendil-works/pi-ai/compat";
 import type { AssistantMessage } from "@earendil-works/pi-ai";
 import type { ExtensionContext } from "@earendil-works/pi-coding-agent";
 import type {
@@ -151,10 +151,19 @@ async function runOnce(
       return { kind: "auth", message: authMessage };
     }
 
+    // Mirror the main loop (model-runtime stream): auth resolution can carry a
+    // seat-specific baseUrl (e.g. GitHub Copilot business/enterprise endpoints).
+    // The shipped model data pins the individual host, which 421s other seats,
+    // so the resolved auth baseUrl must win over the static model baseUrl.
+    const providerAuth = await ctx.modelRegistry.getProviderAuth(model.provider);
+    const effectiveModel = providerAuth?.auth.baseUrl
+      ? { ...model, baseUrl: providerAuth.auth.baseUrl }
+      : model;
+
     // Pass the combined signal so the underlying fetch is cancelled immediately
     // either when the user presses Esc, or when an idle/ceiling timeout fires.
     const responseStream = stream(
-      model,
+      effectiveModel,
       {
         messages: [
           {